The present invention relates to antigens for the prevention/treatment/suppression of Clostridium difficile infection (CDI). Also provided are methods for generating said antigens, methods for generating antibodies that bind to said antigens, and the use of said antibodies for the prevention/treatment/suppression of CDI.
Clostridium difficile infection (CDI) is now a major problem in hospitals worldwide. The bacterium causes nosocomial, antibiotic-associated disease which manifests itself in several forms ranging from mild self-limiting diarrhoea to potentially life-threatening, severe colitis. Elderly patients are most at risk from these potentially life-threatening diseases and incidents of CDI have increased dramatically over the last 10 years. In 2010 in the UK there were over 21,000 cases of CDI with over 2,700 associated deaths. CDI costs the UK National Health Service in excess of £500M per annum.
The various strains of C. difficile may be classified by a number of methods. One of the most commonly used is polymerase chain reaction (PCR) ribotyping in which PCR is used to amplify the 16S-23S rRNA gene intergenic spacer region of C. difficile. Reaction products from this provide characteristic band patterns identifying the bacterial ribotype of isolates. Toxinotyping is another typing method in which the restriction patterns derived from DNA coding for the C. difficile toxins are used to identify strain toxinotype. The differences in restriction patterns observed between toxin genes of different strains are also indicative of sequence variation within the C. difficile toxin family. For example, there is an approximate 13% sequence difference with the C-terminal 60 kDa region of toxinotype 0 Toxin B compared to the same region in toxinotype III Toxin B.
Strains of C. difficile produce a variety of virulence factors, notable among which are several protein toxins: Toxin A, Toxin B and, in some strains, a binary toxin which is similar to Clostridium perfringens iota toxin. Toxin A is a large protein cytotoxin/enterotoxin which plays a role in the pathology of infection and may influence the gut colonisation process. Outbreaks of CDI have been reported with Toxin A-negative/Toxin B-positive strains, which indicates that Toxin B is also capable of playing a key role in the disease pathology.
The genetic sequences encoding Toxin A and Toxin B (Mw 308k and Mw 269k, respectively) are known—see, for example, Moncrief et al. (1997) Infect. Immun. 63:1105-1108. The two toxins have high sequence homology and are believed to have arisen from gene duplication. The toxins also share a common structure (see FIG. 1), namely an N-terminal glucosyl transferase domain, a central hydrophobic region, four conserved cysteines, and a long series of C-terminal repeating units (RUs).
Both Toxins A and B exert their mechanisms of action via multi-step mechanisms, which include binding to receptors on the cell surface, internalisation followed by translocation and release of the effector domain into the cell cytosol, and finally intracellular action. Said mechanism of action involves the inactivation of small GTPases of the Rho family. In this regard, the toxins catalyse the transfer of a glucose moiety (from UDP-glucose) onto an amino residue of the Rho protein. Toxins A and B also contain a second enzyme activity in the form of a cysteine protease, which appears to play a role in the release of the effector domain into the cytosol after translocation. The C. difficile binary toxin modifies cell actin by a mechanism which involves the transfer of an ADP-ribose moiety from NAD onto its target protein.
Current therapies for the treatment of C. difficile infection rely on the use of antibiotics, notably metronidazole and vancomycin. However, these antibiotics are not effective in all cases and 20-30% of patients suffer relapse of the disease. Of major concern is the appearance in the UK of more virulent strains, which were first identified in Canada in 2002. These strains, which include those belonging to PCR ribotype 027 and toxinotype III, cause CDI with a directly attributable mortality more than 3-fold that observed previously.
New therapeutics are therefore required especially urgently since the efficacy of current antibiotics appears to be decreasing.
One approach is the use of antibodies which bind to and neutralise the activity of Toxin A and/or Toxin B. This is based on the knowledge that strains of C. difficile that do not release these toxins, so called non-toxigenic strains, do not cause CDI. By way of example, animals can be immunised, their sera collected and the antibodies purified for administration to patients—this is defined as passive immunisation. In another approach patients with CDI or subjects at risk of developing such infections can be immunised with antigens which result in an increase in circulating and mucosal antibodies directed against Toxin A and/or Toxin B—this is defined as active immunisation.
A critical requirement for both active and passive immunisation is the availability of suitable antigens with which to immunise the patient or animal respectively. These can comprise the natural toxins which can be purified from the media in which suitable toxigenic strains of C. difficile have been cultured. There are several disadvantages to this approach. Both Toxin A and Toxin B are present in culture medium in only small amounts and are difficult to purify without incurring significant losses. Thus, it is both costly and difficult to obtain the amounts necessary to meet world-wide needs. In addition, the natural toxins are unstable and toxic.
The above mentioned problems have resulted in there being few available C. difficile vaccine candidates. To date, the only CDI vaccine in late-stage development is based on a mixture of native (i.e. naturally occurring) Toxins A and B, which have been extensively inactivated by chemical modification (Salnikova et al. 2008, J. Pharm. Sci. 97:3735-3752).
One alternative to the use of natural toxins (and their toxoids) involves the design, development and use of recombinant fragments derived from Toxins A and B. Examples of existing antigens intended for use in treating/preventing a C. difficile infection include peptides based on the C-terminal repeating units (RUs) of Toxin A or Toxin B—see, for example, WO 00/61762. A problem with such antigens, however, is that they are either poorly immunogenic (i.e. the antigens produce poor antibody titres), or, where higher antibody titres are produced, the antibodies demonstrate poor neutralising efficacy against C. difficile cytotoxic activity (i.e. insufficient neutralising antibodies are produced).
There is therefore a need in the art for new vaccines/therapies/therapeutics capable of specifically addressing C. difficile infection (CDI). This need is addressed by the present invention, which solves one or more of the above-mentioned problems.